Currently, Esprima uses large script-generated regular expression patterns to match identifier characters correctly:

Now that Unicode RegExp property escapes are now part of the language, we could use \p{ID_Start} and \p{ID_Continue} instead to massively simplify the Esprima source code and get rid of separate script + manual updating workflow.

Pros:

• Source code would become simpler and more readable
• No more need for a separate build script + manual updating workflow

Cons:

• IIUC, TypeScript doesn't currently support transpiling Unicode RegExp property escapes, so we might need to add a transpilation step to the existing build process.

Thoughts?